385
A Geophysical Survey at Dun Ailinne,
County !Gldare, Ireland
Susan A. Johnston
George Washington University
Washington, D. C.
Douglas Campana
U. S. National Park Service (Retired)
Pam Crabtree
New York University
New York, New York
Dun Ailinne) in County I(ildare) was an important center of ceremonial and ritual activity in the Irish Iron Age (600 B.G-A.D. 400). The site is located on a hill and is surrounded
by an earthen bank and ditch. Excavation of the summit 35 years ago revealed a series of
timber structures that were a1JJuablythe focus of the site)sIron Age use. This early research
examined only ca. 10% of the site)s interio1; sofrom 2006 to 2008 a new program of
research employed a ta1JJetedtopographic survey and a magnetometer survey to determine if
there was evidence of additional ftatures inside the bank and ditch) or whether the rest of
the site beyond the summit was empty. The results show that the area outside of the excavated summit was characterized by many and varied ftatures of archaeological interest)
including additional enclosuresand likely structures of the Iron Age and both earlier and
later periods. These data allow us to better understand the use of Dun Ailinne during its
I ron Age florescence and provide us with directions for future archaeological research.
Introduction
The site of Dun Ailinne, County IZildare, Ireland, is enclosed by an earthen bank.and ditch and located on the top
of I<nockaulin Hill (FIG. I). It is one of the major sites of
the Irish Iron Age (ca. 600 B.C.-A.D. 400). In documentary
sources of the Early Christian period (A.D. 400-800), Dun
Ailinne is consistently grouped with three other sites (Tara,
County Meath; N avan Fort, County Armagh; and
Rathcroghan, County Roscommon) (FIG. r) as significant
pre-Christian social and political centers (Grabowski
1990; Hicks 2007). Together, these four sites have become
known conventionally as the "royal sites" of the Irish Iron
Age, since early historical documents peopled them with
semi-legendary Icings.The use of the term royal site is retained here despite the fact that "royal" conjures an inaccurate image of the social reality during that period.
Recent geophysical research and excavation (Newman
1997; Lynn 1997; Johnston and Wailes 2007; Barton and
Fenwick 2005) demonstrated that the historical sources,
while misinterpreting the social context of the royal sites,
were correct in recognizing them as a coherent group. The
sites differ in significant ways, however. They share a set of
features, including a hilltop location in a landscape of ritual sites from all prehistoric periods, an outer bank with an
internal ditch surrounding the site, and the presence of
timber structures including both circular and figure-ofeight shapes within the bank. and ditch enclosure. Among
their differences, N avan and Tara both have central
mounds, which Dun Ailinne lacks, Tara incorporates an
earlier burial in its enclosure, N avan has the extraordinary
40 m-Iong structure, and all differ in size and internal organization. Nevertheless, the shared characteristics would
have allowed an ancient observer to know that they were
in a royal site; the differences presumably reflected regional and local preferences for the structuring of space and associated behaviors (Johnston 2006). As research continues
(Waddell, Fenwick, and Barton 2009), interpretations of
the royal sites become more sophisticated, contributing to
our understanding of ceremonial life in the Irish Iron Age.
Along with the excavations at N avan and Tara (Lynn
1997; O'Sullivan 2005; Grogan 2009), Bernard Wailes'
386
Geophysical Survey at Dun Ailinne)
IrelandlJohnston)
Campana)
Belfast
and Crabtree
other structures, and if so, what kinds were they and where
were they located; or was it left open, perhaps to facilitate
large gatherings or for other ritual reasons; and how is it
similar to and different from the other royal sites? In order
to address these questions, we undertook a geophysical
and topographic survey of the interior of the site. This project included both remote sensing (resistivity and magnetometry) and a targeted topographic survey of several areas
within the bank and ditch.
Geology, Geomorphology,
Rathcroghan.
Dublin.
Dun Ailinne
*
Ireland
Figure 1. Map showing the location of Dun Ailinne. The Irish Ordnance Survey Grid Reference for Dun Ailitme is N818078 and the
Irish National Grid is £282030, N207880. The other three royal
sites are also indicated for reference.
excavations at Dun Ailinne from 1968 to 1975 (Johnston
and Wailes 2007) provided important infonnation for our
understanding of royal sites. Although activity on the hill
began in the Neolithic, during the third millennium B.C.,
and continued into the Bronze Age, Dun Ailinne's major
florescence is represented by large ceremonial structures
that probably date to the early centuries A.D. Iron Age use
of the site culminated in a period characterized by largescale ceremonial feasting (Crabtree 2007). While some of
these sites, most notably Tara, retained significance (particularly in the political realm), others, like DlinAilinne, were
largely eclipsed as ceremonial and ritual centers after the arrival of Christianity in Ireland in the 5th century A.D.
While the earlier excavations at Dun Ailinne provide evidence for site use, they were concentrated primarily on the
summit of the hill. A large area was excavated; but the site
covers some 13 ha, and only about 10% of the site inside
the bank and ditch was tested. This raises a series of questions about how the rest of the site was used: are there
and Topography
The site of Dun Ailinne lies on a prominent hill just
south of IZilcullen, County IZildare (FIG. 2). The hill is
steep, rising to a height of 180 masl, and the bank and
ditch surrounding the site are arranged somewhat off-center, extending more towards the north than the south (FIG.
3). The hill commands panoramic views of the surrounding
landscape, including the central lowlands to the south and
west and the flat pasturelands of the Curragh to the NW.
The River Liffey Basin/Dublin Bay area is clearly visible to
the north, while the eastern horizon is dominated by the
granite uplands of the Wicldow Mountains. With the exception of several low, rounded hills, such as those at Old
I<ilcullen (179 masl, 1.5 km to the south) and Carrighill
Upper (166 masl, 2.5 km to the sW), the surrounding
countryside is characterized by a relatively flat to undulating topography.
Geologically, the hill is composed of slate and graywacke
bedrock
that lies oblique
to the surface, running
ENE-WSW, and that frequently outcrops in rougWy parallel ridges. Its hardness varies; in some places it is very hard,
while in others it is quite soft and friable. This stone was
quarried in the past, as shown by numerous quarry holes
and ditches of various sizes on the inside edge of the enclosure ditch. A sheet of glacial till lies over the bedrock,
and is 1 m or more deep between the ridges of bedrock, but
thin or nonexistent where the bedrock ridges occur. As is
commonly the case, the glacial till contains fragments of diverse geological origin (including Antrim flint) and varies
considerably in texture; in most places it is more like gravel but in others it is fine and sandy (Johnston and Wailes
2007: 6). While specific effects of this geology have not
been established, it is possible that some of the magnetic
anomalies detected in the geophysical survey are the result
of the varied constituent elements of the underlying rock,
as well as its proximity to the surface.
Mter the site was abandoned as a ceremonial center, the
hill was used for agricultural purposes during different
time periods. Both the topographic and magnetometer
surveys described below provide evidence that the site was
plowed on multiple occasions, and the field boundaries vis-
Journal of Field Archaeology;Vol. 34) 2009
387
Figure 2. DUn Ailinne, looking NW. The bank and ditch surrounding the site are marked by the first dark
line of vegetation coming down from the summit of the hill. Photograph by S. A. Johnston.
ible in earlier aerial photos of the site attest to previous
agricultural fields. Since the 1950s the site has been under
improved grassland and most of it is cleared. The exception
is an area to the sw, where dense gorse reappeared after
some earlier clearance. The hill is now used as pasture for
both cattle and sheep.
The landscape aound Dun Ailinne contains numerous
ritual sites, some visible from the hill itself In a recent survey of the Curragh, the flat open plain that lies to the NW,
Clancy (2006) identified 179 of these ritual sites, the vast
majority of them burial mounds probably representing various prehistoric and historic periods. The lack of dating for
many of these sites mal(es it difficult to determine the sequence of land use in this area. It is likely that some of these
ritual monuments influenced the locations of others.
Summary of the 1968-1975
Excavations
There are few surface indications of the prehistoric use
of Dun Ailinne (FIG. 4). The site is enclosed by a large bank
with a ditch adjacent to its interior side. Such an arrangement is referred to as "hengiform:' since it resembles the
henge monuments (large ceremonial enclosures with varying internal arrangements of structures) of the Neolithic
and early Bronze Age (Hicks 1975; Stout 1991) rather
than the typical arrangement of later hillforts (Raftery
1994). The original entrance through the bank and ditch is
on the east. Within this enclosure, there are two surface features. The first is an arc-shaped embankment on the summit whose age and function could not be determined, but
is certainly post-Iron Age based on excavation. The other is
a rocky depression in the northern part of the hill that was
apparently enlarged sometime in the past. It is known as
St. John's Well, presumably because it retains water when
it rains or because there is (or was) an underground spring
there (although this has not been determined).
Excavations on the summit of the hill were initiated in
1968 (see Johnston and Wailes 2007 for a detailed analysis of the excavations). They revealed that, while intensive
388
GeophysicalSurvey at Dun Ailinne) IrelandlJohnston) Campana) and Crabtree
Figure 3. Aerial photograph of Dun Ailinne, looking sw, taken in 2007. Photograph courtesy of Frank
Coyne, Aegis Archaeology Ltd.
use of DUn Ailinne occurred during the Iron Age, there
was activity on the hill beginning in the Neolithic. Remains
dating to the first half of the third millennium B.C. consisted of a single disturbed burial (Feature 293) and scattered lithic and ceramic artifacts, as well as a circular ditch
that probably dates to this period. Evidence from the
Bronze Age, perhaps a thousand years later, was ephemeral: a single ceramic vessel, the sole survivor of a disturbed
burial (Feature 2790), was the only definite Bronze Age artifact, though some of the other artifacts, including lithics,
beads, and metal jewelry could also belong to the Bronze
Age.
In contrast, the evidence from the Iron Age was substantial. A sequence of at least three circular timber complexes was built on the summit of the hill, each representing a separate phase (FIG. 5). Using the labeling conventions of the original excavator, these were designated by
colors (see Johnston and Wailes 2007: 6 for an explanation
of this system). The first major structure, the smallest and
simplest of the three, was the White phase. It comprised a
single circular palisade trench 23 m in diameter with a fea-
ture at the center composed of two opposed arcs of postholes. This was succeeded by the Rose phase, a more complex undertalang. The Rose phase structure had a figure-ofeight shape, with a smaller enclosure (21 m maximum diameter) adjacent to the southern perimeter of a larger enclosure (38 m maximum diameter), both marked by multiple concentric timber palisades. The entrance to the latter
was elaborated by the construction of a funnel-shaped timber palisade that narrowed as it approached the larger circle. The final and most elaborate of the three complexes
was the Mauve phase. This comprised a series of concentric circular structures. The outermost was a double palisade (43 m maximum diameter), with an entrance to the
NE. Within this was an open timber circle (20 m diameter)
made of 3 a large posts surrounding a smaller circular feature (6 m diameter). This last small circular feature had no
apparent entrance, the timber palisade that marked it being
continuous. The final phase of Iron Age use, as noted
above, mostly consisted of faunal remains that showed evidence of butchering and roasting. These were interpreted
as evidence of large-scale feasting (Crabtree 2007).
Journal ofFieldArchaeologyjVol. 34) 2009
389
a:::~~:~
not fully
excavated
---... -•...•
N
1
I
Figure 4. Site plan of Dun Ailinne from the 1968-1975 excavations showing the areas excavated. The
outlined area in the center is the summit of the hill, and a detail of this is shown in Figure 5. Adapted
from Johnston and Wailes 2007, fig. 1-1.
The timber structures were probably used for ritual or
ceremonial purposes rather than residential ones. They are
generally too large to have been roofed without internal
support, for which there is no evidence, nor is there evidence of hearths or other features typical of domestic use.
While there is animal bone, it is almost exclusively confined
to the final phase of Iron Age use and not associated with
any of the previous timber structures that appear to have
been dismantled before the animal bone was deposited.
Thus, the lack of domestic debris or features associated
with these structures, suggests that they are ritual or ceremonial in nature.
Structural evidence from the Iron Age was extensive,
but the quantity of artifacts from the period was surprisingly small. Most artifacts typical of the Irish Iron Age were
represented (e.g., glass beads and bracelets, bronze rings
and brooches, and iron weapons), but there were few examples of each. Of note are two largely intact bronze fibulae, a bone ornament (possibly a pin head), an iron spearhead, and a sword. Overall, the recovered artifacts suggest
the objects were brought to and/or worn at the ceremonial gatherings that arguably characterized Iron Age Dun
Ailinne. In addition, a number of artifacts (e.g., slag,
worked bone fragments, and possibly waste glass) indicate
low-level manufacturing, possibly associated with the larger ceremonIes.
390
Geophysical Survey at Dun Ailinne) IrelandIJohnston) Campana) and Crabtree
50000
45000
o
'l.o.I'U'U'u'u'
__
meters
20
...&..--_......I-_~,
- 50000
2790
. .-..;... -.
t'
~
.
.•
Other features
- 45000
Neolithic trench 281
Figure 5. The main excavated area on the summit of the hill showing the major features by phase. This is
the outlined area indicated in Figure 4. Adapted from Johnston and Wailes 2007, fig. 2-1.
The Survey
The excavations permitted the interpretation of the Iron
Age use of Dun Ailinne, although they were confined to a
limited area within the bank and ditch on the summit of
the hill. This meant that some features could not be fully
defined (e.g., the extent of the funnel entrance to the Rose
phase structure), and limited information was available
about the remainder of the site. A magnetometer survey
over the interior was carried out in 1968 (Ralph 1968),
but no anomalies of particular interest apart from generalized areas of burning were identified (Johnston and Wailes
2007: 4). It is not entirely clear why this was so, given the
wealth of features identified some 40 years later, but it may
have to do with aspects of geophysical research of that time
period.
In order to expand our understanding of activity at DUn
Ailinne beyond the earlier excavated area, we initiated a
geophysical and topographic survey of the site's interior
over three seasons, from 2006 to 2008. The survey grid
was set out with a total station using a previously established control network consisting of four temporary survey
stations, and is based on 20 m multiples of the Irish National Grid. The primary focus was a magnetometer survey
of the 13 ha within the bank and ditch. With the exception
of two areas in the sw of the site (about 1.5 ha in total) that
were covered in thick gorse, the entire area inside the bank
and ditch was surveyed, as well as an area just outside the
site's original entrance. In the first season, we also conducted a resistivity survey over a small area on the summit,
but the results, while interesting, did not warrant further
use of this technique given its time consuming and expen-
Journal of Field Archaeo IogyfVo I. 34y 2009
sive nature. Finally, we conducted a targeted topographic
survey in three interior places and across the bank and ditch
in four locations.
The topographic survey was conducted using a Topcon
GTS-235W total station in conjunction with a TDS-Recon
data collector with TDS Survey Pro software. All data were
transferred to Excel spreadsheets and analyzed using Surfer
8 software to produce a variety of graphic representations.
The accuracy of the measurements depends on both the instrument and the precision of its operators. The instrument
documentation registers 2 mm accuracy but under actual
conditions it is more likely to be within 1-2 cm horizontally and vertically.This approach produced data that could
be presented in several different formats depending on the
specific conditions for each area, and that could also be
combined with the data from the magnetometer survey.
The magnetometer survey was conducted using two
Geoscan FM36 fluxgate gradiometers. Measurements
were recorded at 0.5 m intervals moving south to north
along parallel traverses spaced 0.5 m apart within a grid
composed of 20 x 10 m survey panels. The survey of the
interior had maximum dimensions of 440 m N-S x 356 m
E-W, and encompassed a total area of approximately 11.5
ha. In addition, a small, subrectangular area (maximum dimens ions 100 m N-S x 63 m E-W) just outside the original enclosure entrance was investigated. Data collected
during the survey were downloaded and processed using
Geoplot software. The survey panels have been corrected
for drift and edge-matched, and excessively high readings
were eliminated from the combined data to reduce the effect of spuriously high magnetic anomalies ("spikes") due
to ferrous litter, etc., in an attempt to highlight weaker
anomalies of potential archaeological significance.
391
trance down the hill for about 40 m (the "outer roadway").
The surface indications of the inner roadway petered out
well before it reached the summit area that had been excavated earlier and it was uncertain how the roadway related
to the series of timber structures subsequently identified.
The topographic survey was initiated to trace the surface
indications of this roadway nearer to the summit. Second,
we wanted to get a precise topographic image of St. John's
Well. It was not investigated in the earlier excavations, but
since wet places were often given ritual significance during
the Iron Age in Ireland (Raftery 1994; Cooney and Grogan 1999), we attempted to obtain data on the feature as a
precursor to future investigations.
Inner Roadway
The area over the inner roadway was surveyed as a series
of parallel transects perpendicular to the length of the road.
Readings along each transect were tal(en at 50 cm intervals,
with a spacing of 1 m between transects at the east end of
the road, increased to 5 m near the central portion, and reduced to 2 m intervals at the western end. Cross sections
of the roadway generated by the survey indicate that, while
shallow and indistinct, there is a measurable depression as
the roadway nears the summit, a subtle feature almost invisible to the naked eye. Figure 7 (top) shows a topographic image of the roadway superimposed on the data
from the magnetometer survey. The roadway runs straight
up from the site entrance through the apparent entrance to
the summit enclosure (FIG. 7, bottom). This was not obvious because the depression in the ground is not visible
along its entire length and there is no surface indication of
the summit enclosure or its apparent entrance. Thus, the
survey has confirmed the projected path of the roadway.
Topographic Survey Results
St. ]ohn)s Well
The topographic survey provided significant data on the
profile of the bank and ditch and on the general surface
characteristics in those areas where it was employed. For
example, on the summit it revealed evidence of later historical agricultural use of the site in the form of "ridge and
furrow" marks, a series of undulating, parallel ridges that
represent the last phase of plowing at the site (FIG. 6).
These could date anywhere from the medieval period to
the 19th century, though the close spacing suggests they
are post-medieval. Two other results are also notable. During the excavations, a depression was observed as a surface
feature that ran from the site entrance straight up the hill
towards the summit for about 70 m (Johnston and Wailes
2007: 28-29). This was dubbed the "inner roadway" to
distinguish it from a second, more sinuous, shorter, and
shallower depression that extended from outside the en-
St. John's Well has potential significance in the context
of known Iron Age ritual activity in Ireland and elsewhere
in Britain and Europe. A major focus of ritual life ,vas the
deposition of objects (most commonly metal, e.g.,
weapons, jewelry, and harness fittings, though other, more
perishable objects may also have been included) in watery
places such as lakes, rivers, and bogs (Raftery 1994;
Cooney and Grogan 1999). There is no evidence to indicate when St John's Well was enlarged, yet an Iron Age date
is a viable hypothesis since this was the period when the hill
received the most attention in prehistory. The well was surveyed over an approximately 12 x 12 m grid, with readings taken at 50 cm intervals, and at closer intervals in
areas with higWy variable topographic relief. Topographic
and surface maps were produced by interpolating data to
10 cm intervals using the Surfer kriging algorithm.
392
Geophysical Survey at Dun Ailinne)
IrelandIJohnston)
Campana)
and Crabtree
Figure 6. Surface topographic image of the summit of the hill. The inset shows the location of the surveyed area relative to the site plan. The semicircular mound corresponds with O'Donovan's summit fort
in the site plan in Figure 4. The remains of ridge-and-furrow plowing are dearly visible as undulating
ridges south of the mound, running approximately NNE-SSW across the area. Grid references are provided to indicate the area's location relative to the Irish National Grid.
The surface image (FIG. 8, left) shows the asymmetric
shape of the well's interior surface. It is roughly square, and
three sides of it are fairly steep or near vertical. The fourth
side, to the NW, slopes more, apparently the result of waste
Journal
o
of Field ArchaeologyjVol.
34) 2009
393
100m
1
1
o
,-I
100m
-l-
-"----
I
Figure 7. Topographic image of the inner roadway superimposed on the magnetometer image. Unlabeled
image (top); and image with the entrance to the summit enclosure Inarked with black dots and the center
line of the roadway indicated by a black line (bottom).
material deposited during the widening of the original geological feature. Large projections visible on the surface are
rock outcrops, emerging from the grassy surface on the
sides and at the top of the well. The orientation of the well
394 Geophysical Survey at Dun Ailinne) IrelandIJohnston)
Campana)
corresponds with a possible gap in the sumn1it enclosure as
shown in Figures 9 and 10. The well is marked with a circle in Figure 10, and its southeastern edge is approximately parallel with the enclosure gap.
Owing to the possible importance of the well in prehistory, the well surface image was overlain with the data from
the magnetometer survey (FIG. 8, right). As this image
shows, there is one anomaly lying in the center of the well
and a second one adjacent to the southwestern edge. These
anomalies do not appear in the magnetometer
image
shown in Figure 8 (this page) because of the large scale
necessary to encompass the whole site. With the topographic survey, however, we were able to focus on the well
and examine it in more detail for the placement of magnetic anomalies. The anomaly positioned inside the interior surface of the well is of particular interest. Its significance is uncertain; it may be a pit, either natural or artificial' possibly containing artifacts or other deposits, or it
and Crabtree
may reflect some other human activity associated with the
well. If so, some connection with the Iron Age is possible.
It may also be that the bedrock beneath the surface at the
bottom of the well is registering as a magnetic anomaly as
a result of the constituent elements of the rock. Without
further research it is impossible to know.
Magnetometer Survey Results
Figure 9 shows the magnetometer data for the interior
of the site. A large number of features (indicated by the
darker dots and lines in the grayscale image) are visible,
demonstrating that the interior of the site was indeed used
heavily. Interpreting a magnetometer image is like a combination of data analysis and a Rorschach test, and many of
the features are uncertain. Figure 10 outlines those discussed in the text. One that clearly stands out is the dark,
discontinuous band of very strong magnetic anomalies
(Feature 19; FIG. 10) running ESE-NNW (at one end it ap-
Journal ofFieldArchaeologyjVol. 34) 2009
395
Figure 8. Surface plots of St. John's Well (left) and St. John's Well overlain with the magnetometer data
(right). The dark areas (center and adjacent to swedge) indicate magnetic anomalies. Grid references are
provided to indicate the feature's location relative to the Irish National Grid.
pears as a large, inverted L shape). It is possible that this
band, and in particular the L-shaped anomaly, are of archaeological interest, but their general orientation and extent suggest that they are geological and may relate to an
igneous intrusion or seam in the underlying bedrock. A
number of other features are also well delineated and of
clear archaeological significance. Below, we discuss the
most important features.
More Recent Historical Features
Linear striations over the surface of the site are one of
the more striking aspects of this image (FIG. 9). These are
the remains of plowing, probably dating to the post-medieval period. The overall ridge and furrow pattern indi-
cated by magnetometry is aligned in various directions in
different parts of the site, forming a crisscross pattern in
some areas. The overlapping pattern suggests multiple
phases of plowing, perhaps a result of changing field alignments in different periods. The topographic survey mirrors
the magnetometry data for the most recent ridge and furrow, but the earlier, underlying pattern situated at different
angles may suggest that a long period of time passed between the different phases of plowing. Further indication
of agricultural use lies in the field boundaries (Features
1-4), three of which converge just NW of the center of the
site. These three appear on the Irish Ordnance Survey
maps of the 19th and early 20th centuries, and so date at
least to this period and possibly earlier.
GeophysicalSurvey at Dun Ailinne) IrelandlJohnston) Campana) and Crabtree
396
N
//
1/'
,-
1~···/
\
\
\
\
\
\
\
\
7
nT
-7
200m
Figure 9. Grayscale image of the 2006-2008 gradiometry survey of the interior of Dun Ailinne. Magnetic
anomalies are indicated by darker (positive) or lighter (negative) color.
Iron Age Features
While the dates of many of the features are unknown,
several can be assigned to the Iron Age. One of these is the
series of arcs in the center of the site (collectively designated Feature 5), representing the palisade trenches of the
Rose and Mauve phase timber structures discovered in the
earlier excavations, and dating to the Iron Age (Johnston
and Wailes 2007). Another is the subcircular enclosure that
rings the summit of the hill (Feature 6). It appears as a narrow band of positive magnetic readings 0.5-1 m in width
and is quite large, with maximum overall dimensions of approximately 240 m N-S x 200 m E-W. Given its width and
the fact that the excavated Iron Age trenches held timber
Journal
of Field ArchaeologyjVol.
34) 2009
397
N
St.John's Well
l~
16
\
\
\
\
\
\
\
\
7
nT
-7
200m
Figure 10. Grayscale image as in Figure 9, with major features outlined
in the text.
in black. Numbers
palisades, it, too, probably held a palisade. There are possible indications of a number of gaps along its circumference, but the formal entrance appears as a gap to the east
where the enclosure is almost pointed. This entrance also
aligns \vith that through the bank and ditch and is integrated with the funnel entrance to the Rose phase complex
(Feature 7). For reference, Figure 11 shows the plans of the
are referred to
Rose and Mauve phase timber structures superimposed on
the magnetometer image from Figure 9. This shows that
the funnel begins at the entrance into the larger of the Rose
phase palisades and ends at the entrance of the summit enclosure. The spatial relationship, therefore, indicates that
the summit enclosure is contemporary with the Rose phase
complex, which in turn has been dated to the Iron Age
398
GeophysicalSurvey at Dun Ailinne) IrelandIJohnston) Campana) and Crabtree
measuring about 4 m in diameter, lies at its center. Some
of the other discrete magnetic anomalies visible in its interior may represent pits or other features of archaeological
significance.
Earlier Enclosure
Figure 11. Enlargement of the grayscale image of the area inside the
sUlTImit enclosure with the timber structures frOlTIthe Mauve and
Rose phases superimposed. The funnel entrance of the Rose phase
structure in the grayscale image is integrated with the summit enclosure entrance.
(Johnston and Wailes 2007: 177-182). As noted above,
the inner roadway also extends through this entrance.
The continuation of the Rose phase entrance funnel
(Feature 7) indicated on the magnetometry image is also
Iron Age. The full extent of this feature could not be determined during excavation because of limits on the excavated area. Figure 11 focuses on the magnetometer image
within the summit enclosure (Feature 6), and shows the
superimposed Rose and Mauve phase structures from the
excavation plan (FIG. 5). Here, it can be seen that the funnel, indicated by a pair of narrow bands of positive magnetic readings spaced roughly 30 m apart and running
ENE-SSW, becomes parallel 35-40 m after it begins and
then continues for some 60 m before ending at the entrance to the summit palisade. Similarly, the rest of the arcshaped feature (Feature 8), which is on the western edge of
the excavated area in Figure 5 (labeled as Feature 229), was
also identified. It shows up very distinctly as a positive
n1agnetic anomaly measuring approximately 20 m in diameter. Its narrow width suggests that it n1ay represent a
slot-trench for a timber structure akin to some of the
other excavated features. A small, roughly circular anomaly,
A particularly important feature is the massive enclosure
(Feature 9) which appears to predate the bank and ditch
surrounding the site. It appears as a slender, discontinuous
band of positive magnetic readings that, like the summit
enclosure, is also subcircular in plan; however, this feature,
which is far larger, is oriented on a different axis than the
summit enclosure. More importantly, it not only encompasses most of the area inside the bank and ditch, but also
extends a short distance beyond the eastern entrance, apparently running underneath the bank and ditch. Although
quite clearly defined from the sw around to the east, where
it averages 1 m in width, the rest of this enclosure can
only be traced in places on the north and NW; on the west,
where survey was limited by gorse cover, it exhibits virtually no geomagnetic expression. Sufficient portions have
been geophysically mapped to illustrate its enormous scale,
extending to some 390 m in maximum diameter
(NNE-SSW). Its relatively narrow width and the occurrence
of numerous, irregularly spaced gaps along its circumference suggest that the enclosure may have held a discontinuous timber palisade. There is no way to determine the
date of this feature from geophysical data, but the fact that
it seems to run under the Iron Age bank and ditch suggests
tl1at it predates them. Since there is evidence for the use of
the hill from the Neolithic through the Iron Age (Johnston
and Wailes 2007), it could date to any of these periods.
Features of Uncertain Date
Most of the features visible in tl1e magnetometer image
cannot be dated with precision. Depending on how many
of the faint features are included, the total number of features ranges from dozens into the hundreds. Many of them
are circular, and the narrow width of some suggests that
they might also be slot trenches for palisades. The lllore
clearly delineated are described here. Figure 12 shows enlargements of some of these.
In the northwestern part of the site, there are several arcshaped and circular features expressed as faint, positive
magnetic anomalies, with diameters ranging from about 3
to 6 m. Included in this group is a small, circular feature
(Feature 10; FIG. 12), 5 m in diameter, with a discrete n1agnetic anomaly at its center and two apparent pairs of conjoined or overlapping arcs to the NE (collectively designated Feature 11; FIG. 12). To the south of these is an inverted L shape (Feature 12; FIG. 10) indicated by positive mag-
Journal
of Field ArchaeologyjVol.
34) 2009
399
/,l
N
100 m
Figure 12. The magnetometry
Figure 10.
image with selected areas enlarged. Feature numbers are tb.e same as in
netic readings. It runs in a NNE direction for a distance of
approximately 20 m before curving to the wsw and continuing for a further 10m. Within the sumlnit enclosure,
just to the north of the summit (Feature 13; FIG. 10), is a
feature defined by two closely spaced, concentric rings of
positive magnetic readings approximately 20 m in overall
diameter. This partially overlaps with a discontinuous,
semicircular arc of positive magnetic gradient roughly 15
m in diameter that is either part of the first feature or a sep-
arate feature. Near the apparent entrance to the summit enclosure is a faint, positive magnetic, circular anomaly (Feature 14; FIG. 12). It is 14 m in diameter and appears to be
conjoined to the sw, figure-of-eight style, with a smaller
circular anomaly 5 m in diameter; of particular interest are
the suggestions of a funnel-like entrance feature - defined
by two linear anomalies - at the eastern side of the larger of
the two circular features and the discrete, positive anomaly
at its center that could represent a pit.
400
Geophysical Survey at Dun Ailinne) I relandlJohnston) Campana) and Crabtree
Finally, there is a line of semicircular features just within the southern edge of the summit enclosure (Feature 15;
FIG. 12), each measuring 5-10 m in diameter. The relationship between these and the summit enclosure is unclear, but some of them may run underneath it. These features are of interest because they might be ring ditches and
may contain burials. Burials are present at all four royal
sites; some are in close proximity to the sites, as with the
passage tomb on the hill of Tara (O'Sullivan 2005), and all
known royal sites are situated in landscapes containing
burials from various prehistoric periods (Johnston 2006).
Outside the Entrance
Multiple features were detected to the east of the main
enclosure entrance (FIGS. 9-12). These include as many as
six circular anomalies of positive magnetic readings, the
largest of which (Feature 16) is about 15 m in diameter
and appears to overlap with the line of the earlier enclosure
(Feature 9) to the west. Adjoining, or possibly overlapping
this feature on the east is a second, smaller circular anomaly of positive magnetic readings (Feature 17), approximately 6 m in diameter, with a discrete magnetic anomaly
at its center. It also appears to overlap with the line of the
earlier enclosure, as does another, even smaller circular
anomaly located a short distance to the south. Finally, the
sunken outer roadway, which begins just outside the entrance through the bank and ditch, registers as a pair of parallellinear features of negative magnetic gradient (Feature
18), oriented ENE-WSW and spaced approximately 6 m
apart; these continue down the slope beyond the edge of
the survey area.
Interpretations
Geophysical investigation has become a standard tool in
archaeological research and has provided important data
on Irish royal sites. The work undertaken at Tara (Newman
1997) and by various researchers at Navan (e.g., I(vamme
1996) reinforces the value of this approach for identifying
characteristic features. Geophysical investigation has also
contributed significant data at Tara (Fenwick and Newman
2002) and Rathcroghan (Barton and Fenwick 2005). To
some extent, these results prompted our research at Dun
Ailinne, and the number of new features identified here has
enhanced our ability to understand its internal structure
more fully and integrate our fmdings with other royal sites.
At least some of the features at DUn Ailinne are Iron Age
or earlier in date. The summit palisade and the continuation of the Rose phase entrance funnel are obviously conneeted with the other known features on the summit and
belong to the Rose phase ceremonial use of the hill in the
Iron Age. The massive enclosure that appears to run be-
neath the bank and ditch can be similarly dated to a time
before the latter were constructed. Radiocarbon dating
suggests that the bank and ditch belong to the later part of
the first millennium B.C. (Johnston and Wailes 2007: 29),
indicating an earlier date for the enclosure, perhaps earlier
in the Iron Age or in the Bronze Age or Neolithic. Similar
enclosures were identified at Rathcroghan, encircling the
mound itself and its associated features (Barton and Fenwick 2005; Waddell, Fenwick, and Barton 2009), and at
Tara, encompassing the Raith na Senad and a Neolithic
passage tomb (Fenwick and Newman 2002). Each of the
three enclosures has a different configuration, location
within the site, and relationship to associated features, and
they may well date to different prehistoric periods. In a
more general sense, the enclosures represent another feature shared by royal sites.
The rest of the features at Dun Ailinne are undated. The
fact that many of them are circular suggests contemporaneity with the known Iron Age features on the summit,
but circular features in Ireland are hardly confined to the
Iron Age. Several may run underneath the summit enclosure, indicating that they are earlier than this feature. What
is apparent, however, is that the space inside the bank and
ditch at DUn Ailinne held many structures at any given
time and rarely, if ever, was left free of buildings. This is a
shared characteristic of the royal sites. Geophysical research
at three of the other royal sites revealed that their internal
space was far from empty, and likely included structures
from different periods, each phase possibly building on
previous arrangements and incorporating them in symbolically important ways. Analyses of the ways these structures
might have been experienced have been proposed for both
Tara (Newman 2007) and Navan (Lynn 2003). At DUn
Ailinne, this may have been expanded to include the space
outside the bank and ditch. The survey outside the original
entrance showed a number of features that may have been
linked with those inside in ways we are only beginning to
understand.
The structure of internal space at DUn Ailinne has implications for the activities associated with the site. Excavation indicates that DUn Ailinne was used for ceremonial
purposes (religious, economic, political, or some combination of these) that may have changed over time. Rising
above the surrounding landscape, the timber structures visible on the summit would have been an imposing sight,
malcing a symbolic statement associated with these ceremonial activities. As the visual focal point of the hill, it is
possible that the summit alone was the focus of activity,
with the rest of the area inside the bank and ditch remaining open, perhaps forming a buffer between those on the
summit and those outside the banle and ditch. Our geo-
Journal of Field ArchaeologyjVol. 34) 2009
physical survey, however, indicates that this is not correct,
and that the space inside the bank and ditch was at times
filled with structures.
Given the chronological uncertainties, we cannot be certain what was visible on the surface during any given peri0d; we can only speculate about the rituals that might have
occurred. For example, the anomaly at the center of St.
John's Well may be a geological feature, yet there is a~ple
evidence for similar ritual deposits in Irish prehistory, including those associated with N avan. The feature known as
the I<ing's Stables, generally considered to be part of the
larger N avan complex, is an artificially constructed pond
dating to the later Bronze Age. It is located NE of N avan
Fort and contained an incomplete human skull (Lynn
1977) . Just east of this is Loughnashade, a small natural
lake from which four large bronze horns were recovered in
the 19th century (Raftery 1987). In this larger context,
such deposits at Dun Ailinne may also be associated with
ritual activities.
Burial ritual was also practiced at Dun Ailinne. Excavations revealed two burials, one from the Neolithic and one
from the early Bronze Age (Johnston and Wailes 2007: 76,
82-83), and there are additional burials in the surrounding
landscape (Clancy 2006). Burials are associated with the
other royal sites as well. The passage tomb at Tara (O'Sullivan 2005) containing burials from the Neolithic and the
Bronze Age has already been noted and there are many
probable burials in the surrounding landscape there and at
Rathcroghan (Newman 1997; Waddell 1983, 1988; Waddell, Fenwick, and Barton 2009). Prehistoric burials are
also recorded from the surrounding area at N avan and possibly within the enclosure as well (Warner 1986, 1994;
Lynn 1997). If the circular features at the southern end of
the summit enclosure at DUn Ailinne are ring ditches and
do contain burials, the role of burial ritual here would be
greatly expanded.
Other rituals may have involved movement through the
site, choreographed to incorporate whatever structures,
past and present, were visible at the time. As noted, other
researchers have explored similar possibilities for N avan
and Tara. At Dun Ailinne, these rituals could have integrated the outer and inner roadways, the Rose phase funnel-shaped entrance, and the concentric arrangement of enclosures beginning with the bank and ditch. Depending on
the phase, ritual movement would have included the summit enclosure and the figure-of-eight structure of the Rose
phase or the three concentric timber features of the Mauve
phase. The incorporation of still-visible earlier structures
would have enhanced such ceremonial movement, allowing for the addition of further layers of meaning. Given the
number of features identified in this survey, the possibili-
401
ties for ritual expression at Dun Ailinne were only limited
by the imagination of those who constructed and experienced the ceremonies on the hill.
Conclusion
The geophysical survey at Dun Ailinne has contributed
significantly to the interpretation of this important Irish archaeological site. It has provided important evidence
through the mapping and identification of features and the
combination of these two sources of data. It has also indicated fruitful areas for future research, e.g., determining
which of the myriad anomalies indicated by the magnetometry are natural and which are of archaeological significance, and whether dating evidence for the latter can be
obtained. Specific questions include the following: what
was the anomaly at the center of St. John's Well? Which of
the apparent gaps in the summit enclosure are intentional
and which are the result of the fill(s) in certain areas being
insufficiently magnetic to be detected by the gradiometer?
What is the exact relationship between the bank and ditch
and the earlier enclosure on the hill? These are questions
that can now be addressed given the data provided by the
geophysical survey, and their answers will add significantly
to our understanding of the royal site at Dun Ailinne and
the ceremonial life of Ireland's prehistoric inhabitants.
Acknowledgments
We would like to give our great thanks to the many organizations and individuals involved in this project.
Roseanne Schot and Gerard Dowling took the lead role in
executing the magnetometer survey, and we would like to
thank them for their professionalism, skill, and overall
good humor. Much of the technical detail and some of the
interpretive observations presented are taken from their report. We would also like to thank the Heritage Council,
Ireland, the National Geographic Society, and New York
University for their generous financial support, and NUl
Galway, Ireland, and New York University for providing
the survey equipment. Frank Coyne, Aegis Archaeology
Ltd., provided the aerial photograph of the site, for which
we are most grateful. Three anonymous reviewers provided very useful observations on the text, and we thank them
for their comments. Our warmest gratitude goes to the
many individuals who provided help and support in a variety of ways: the Thompson family of IZnockaulin, I<ilcullen, who kindly facilitated the work on their land;
Robert Shaw (the Discovery Programme) and Conor
Newman (NUl Galway) who established the initial survey
grid; Joe Fenwick and the Department of Archaeology,
NUl Galway, who provided technical support; Zenobie
Garrett, whose tireless and good-humored assistance in the
402
GeophysicalSurvey at Dun Ailinne) IrelandIJohnston) Campana) and Crabtree
field was invaluable; and the people of I<ilcullenand the
IGlcullen Heritage Group, particularly Nessa Dunlea,
Desmond Travers, IGeran Forde, and Brian Byrne (A I(ilcullen Diary), whose warm hospitality, energy, and tireless
devotion to promoting public awareness of DUn Ailinne
and the work carried out there over many years make it
such fun to work there.
SusanA. Johnston (ph.D. 1989) University of Pennsylvania)
is an Associate Professorial Lecturer at Ge01;geWashington
University. Her research interests include the archaeological
analysis of ritual sites)predominantly ftcused on Irish prehistory. Mailing address: Department ofAnthropology) Geo1lJe
Washington University) 2110 G St. ~
Washington D. C.
20052. E-mail: sjohnsto@gwu.edu
Douglas K Campana (ph.D. 1981) Columbia University)
retired from the U. S. National Park Service in 2006. His research interests include scientific and computer applications in
archaeology and bone technology.
Pam J Crabtree (ph.D. 1982) University of Pennsylvania)
is an Associate Professor ofAnthropology at New YOrk University. Her research interests include zooarchaeology and the archaeology of later prehistoric and early medieval Europe.
Barton, Kevin, and Joe Fenwick
2005 "Geophysical Investigations at the Ancient Royal Site of
Rathcroghan, County Roscommon, Ireland;' Archaeological Prospection 12: 3-18.
Clancy, Padraig
2006 "The Curragh: A Prehistoric Landscape;' in William Nolan
and Thomas McGrath, eds., I(ildare: History and Society:
Interdisciplinary Essays on the History of an Irish County.
Dublin: Geography Publications.
Cooney, Gabriel, and Eoin Grogan
1999 Irish Prehistory:A Social Perspective.Dublin: Wordwell.
Crabtree, Pam J.
2007 "Biological Remains;' in
Wailes, eds., Dun Ailinne:
1968-1975. Philadelphia:
seum of Archaeology and
Susan A. Johnston and Bernard
Excavations at an Irish Royal Site
University of Pennsylvania MuAnthropology, 157-169.
Fenwick, Joe, and Conor Newman
2002 "Geomagnetic Survey on the Hill of Tara, Co. Meath,
1998-9;' DiscoveryProgramme Reports No.6. Dublin: Royal Irish Academy, 1-17.
Grabowski, Kathryn
1990 "The Historical
32-36.
2007
Johnston, Susan A.
2006 "Revisiting the Irish Royal Sites;' Emania 20: 53-59.
Johnston, Susan A., and Bernard Wailes
2007 Dun Ailinne: Excavations at an Irish Royal Site 1968-1975.
Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology.
Kvamme, Kenneth
1996 ''A Proton Magnetometry
14: 65-88.
7:
Grogan, Eoin
2009 The Rath of the Synods: Tara;,Co.Meath;, Excavations by Sean
p. 6 Riordain. Dublin: Wordwell.
Hicks, Ronald
1975 Some Henges and Hengiform Earthworks in Ireland. Ph.D
dissertation, University of Pennsylvania, Philadelphia. Ann
Arbor: University MicrofIlms.
Survey at Navan Fort;' Emania
Lynn, C. J.
1977 "Trial Excavations at the King's Stables, Tray Townland,
County Armagh;' UlsterJournal ofArchaeology 40: 42-62.
1997
Excavations at Navan Fort 1961-71 by D. M. Waterman.
Belfast: The Stationary Office.
2003
Navan Fort: Archaeology and Myth. Dublin: Wordwell.
Newman, Conor
1997 Tara: An Archaeological Survey. Discovery Programme Monograph 2. Dublin: Discovery Programme.
2007
"Procession and Symbolism at Tara: Analysis of Tech Midchuarta (the "Banqueting Hall") in the Context of the
Sacral Campus;' Oxftrd Journal of Archaeology 26(4):
415-438.
O'Sullivan, Muiris
2005 The Mound of the Hostages;,Tara. Dublin: Wordwell.
Raftery, Barry
1987 "The Loughnashade Horns;' Emania 2: 21-24.
1994
Pagan Celtic Ireland. New York: Thames & Hudson.
Ralph, Elizabeth K.
1968 ''Archaeological
1-2.
Prospecting;'
1WASCA. Newsletter 4(2):
Stout, Geraldine
1991 "Embanked Enclosures of the Boyne Region;' Proceedings
of the Royal Irish Academy 91C: 245-284.
Waddell, John
1983 "Rathcrogan-A
Royal Site in Connacht;' Journal of Irish
Archaeology 1: 21-46.
1988
"Rathcroghan
in Connacht;' Emania 5: 5-18.
Waddell, John, Joseph Fenwick, and Kevin Barton
2009 Rathcroghan: Archaeological and GeophysicalSurvey in a Ritual Landscape. Dublin: Wordwell.
Warner, Richard B.
1986 "Preliminary Schedule of Sites and Stray Finds in the Navan Complex;' Emania 1: 5-9.
1994
Overview of DUn Ailinne;' Emania
"DUn Ailinne's Role in Folklore, Myth and the Sacred
Landscape;' in SusanA. Johnston and Bernard Wailes, eds.,
Dun Ailinne: Excavations at an Irish Royal Site 1968-1975.
Philadelphia: University of Pennsylvania Museum of Archaeology and Anthropology, 183-194.
"The Navan Complex: A New Schedule of Sites and
Finds;' Emania 12: 39-44.